Long-Range Out-of-Sample Properties of Autoregressive Neural Networks

Leoni, Patrick L.

doi:10.1162/neco.2008.09-07-615

Cited by 3 publications

(3 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…What is required, in order to translate our account into neural terms, is an account of how this marginalization operation might be carried out in a neural network. Fortunately, a number of recent theoretical papers have addressed just this problem (Beck & Pouget, 2007; Deneve, 2008; Lee & Mumford, 2003; Litvak & Ullman, 2009; Ma, Beck, Latham, & Pouget, 2006; Pouget, Dayan, & Zemel, 2003; Rao, 2006). One approach that is particularly well suited to the present application was proposed by Rao (2005).…”

Section: Neural Network Modelmentioning

confidence: 99%

Goal-directed decision making as probabilistic inference: A computational framework and potential neural correlates.

Solway¹,

Botvinick²

2012

Psychological Review

172

185

View full text Add to dashboard Cite

Recent work has given rise to the view that reward-based decision making is governed by two key controllers: a habit system, which stores stimulus-response associations shaped by past reward, and a goal-oriented system that selects actions based on their anticipated outcomes. The current literature provides a rich body of computational theory addressing habit formation, centering on temporal-difference learning mechanisms. Less progress has been made toward formalizing the processes involved in goal-directed decision making. We draw on recent work in cognitive neuroscience, animal conditioning, cognitive and developmental psychology and machine learning, to outline a new theory of goal-directed decision making. Our basic proposal is that the brain, within an identifiable network of cortical and subcortical structures, implements a probabilistic generative model of reward, and that goal-directed decision making is effected through Bayesian inversion of this model. We present a set of simulations implementing the account, which address benchmark behavioral and neuroscientific findings, and which give rise to a set of testable predictions. We also discuss the relationship between the proposed framework and other models of decision making, including recent models of perceptual choice, to which our theory bears a direct connection.

show abstract

Section: Neural Network Modelmentioning

confidence: 99%

Goal-directed decision making as probabilistic inference: A computational framework and potential neural correlates.

Solway¹,

Botvinick²

2012

Psychological Review

172

185

View full text Add to dashboard Cite

show abstract

“…The first class of approaches builds on deterministic methods for evaluating exactly or approximately the desired conditional and/or marginal distributions, whereas the second class relies on sampling from the probability distributions in question. Multiple models in the class of deterministic approaches implement algorithms from machine learning called message passing or belief propagation [30]–[33]. By clever reordering of sum and product operators occurring in the evaluation of the desired probabilities, the total number of computation steps are drastically reduced.…”

Section: Introductionmentioning

confidence: 99%

Neural Dynamics as Sampling: A Model for Stochastic Computation in Recurrent Networks of Spiking Neurons

et al. 2011

View full text Add to dashboard Cite

The organization of computations in networks of spiking neurons in the brain is still largely unknown, in particular in view of the inherently stochastic features of their firing activity and the experimentally observed trial-to-trial variability of neural systems in the brain. In principle there exists a powerful computational framework for stochastic computations, probabilistic inference by sampling, which can explain a large number of macroscopic experimental data in neuroscience and cognitive science. But it has turned out to be surprisingly difficult to create a link between these abstract models for stochastic computations and more detailed models of the dynamics of networks of spiking neurons. Here we create such a link and show that under some conditions the stochastic firing activity of networks of spiking neurons can be interpreted as probabilistic inference via Markov chain Monte Carlo (MCMC) sampling. Since common methods for MCMC sampling in distributed systems, such as Gibbs sampling, are inconsistent with the dynamics of spiking neurons, we introduce a different approach based on non-reversible Markov chains that is able to reflect inherent temporal processes of spiking neuronal activity through a suitable choice of random variables. We propose a neural network model and show by a rigorous theoretical analysis that its neural activity implements MCMC sampling of a given distribution, both for the case of discrete and continuous time. This provides a step towards closing the gap between abstract functional models of cortical computation and more detailed models of networks of spiking neurons.

show abstract

“…ARIMA has an in-built mechanism to transform a nonstationary time series into a stationary time series by taking the differencing of the given time series (Brockwell and Lindner 2010). ANNs are asymptotically stationary and it requires the process to be stationary while training the neural network, but when applied to a nonstationary process, the out of sample predictions becomes poor (Leoni 2009). In the hybrid formulation based on ARIMA and ARNN model, this problem can be overcome since we deal with only the additive error terms generated by ARIMA and model it using ARNN model.…”

Section: Asymptotic Stationarity Of the Proposed Modelmentioning

confidence: 99%

Unemployment Rate Forecasting: A Hybrid Approach

Chakraborty

Biswas

et al. 2020

Comput Econ

View full text Add to dashboard Cite

Unemployment has always been a very focused issue causing a nation as a whole to lose its economic and financial contribution. Unemployment rate prediction of a country is a crucial factor for the country's economic and financial growth planning and a challenging job for policymakers. Traditional stochastic time series models, as well as modern nonlinear time series techniques, were employed for unemployment rate forecasting previously. These macroeconomic data sets are mostly nonstationary and nonlinear in nature. Thus, it is atypical to assume that an individual time series forecasting model can generate a white noise error. This paper proposes an integrated approach based on linear and nonlinear models that can predict the unemployment rates more accurately. The proposed hybrid model of the unemployment rate can improve their forecasts by reflecting the unemployment rate's asymmetry. The model's applications are shown using seven unemployment rate data sets from various countries, namely,

show abstract

Long-Range Out-of-Sample Properties of Autoregressive Neural Networks

Cited by 3 publications

References 4 publications

Goal-directed decision making as probabilistic inference: A computational framework and potential neural correlates.

Goal-directed decision making as probabilistic inference: A computational framework and potential neural correlates.

Neural Dynamics as Sampling: A Model for Stochastic Computation in Recurrent Networks of Spiking Neurons

Unemployment Rate Forecasting: A Hybrid Approach

Contact Info

Product

Resources

About